Method for Dynamic Service Orchestration in Fog Computing

Fog computing is meant to deal with the problems which cloud cannot solve alone. As fog closer a user, it can improve some very important QoS characteristics, such as latency and availability. One of challenges in architecture heterogeneous constrained devices dynamic nature end devices, requires service orchestration provide an efficient placement inside nodes. An optimization method needed ensure required level while requiring minimal resources from thus ensuring longest lifecycle whole IoT system. A two-stage multi-objective find best services among available nodes presented this paper. Pareto set non-dominated possible distributions found using integer particle swarm method. Then, analytical hierarchy process used choose distribution according application-specific judgment matrix. illustrative scenario experimental results demonstrate characteristics proposed